Index and search markdown notes with semantic retrieval via LiteLLM embeddings.
npx skills add https://github.com/buyuk-dev/fmql --skill fmql-semantic
Please help me install this Agent Skill. Skill: fmql-semantic Source: https://github.com/buyuk-dev/fmql/tree/main/packages/fmql-semantic/skills/fmql-semantic Command: npx skills add https://github.com/buyuk-dev/fmql --skill fmql-semantic
Enable semantic retrieval over a directory of markdown/frontmatter notes, returning relevant results by meaning rather than exact keywords.
Index a workspace with the semantic backend and then search semantically.
High-intent search queries and answers about installing and using this skill.
You can enable semantic search over markdown notes by running fmql index --backend semantic on your vault directory, then using fmql search --backend semantic to retrieve documents by meaning rather than exact keywords.
Hybrid dense and sparse retrieval combines LiteLLM embeddings with BM25 search to return relevant notes from a vault. This approach ensures robust results by matching both semantic meaning and exact keywords.
Yes, you need a LiteLLM-compatible embedding model and configured API keys via environment variables to index markdown documentation. The semantic backend relies on these to generate vector embeddings for your notes.
You can apply topic clustering and context-aware retrieval across personal knowledge bases using the semantic backend. It processes directories of markdown and frontmatter notes to group related topics meaningfully.
You can improve top-k precision in semantic search results by enabling optional reranking and model pinning. These features refine the initial hybrid retrieval output for higher accuracy in the returned documents.
Semantic retrieval works directly with markdown and frontmatter notes. The indexing process parses frontmatter metadata alongside text content to ensure context-aware search across your workspace.
team-telnyxGenerate text embeddings and perform semantic search via Telnyx AI APIs.
w5851Query OpenAlex for works, authors, and institutions with bibliometric signals.
fratilanicoConfigure dense, sparse, and hybrid document retrieval for RAG pipelines.
zilliztechConvert text into embeddings and retrieve semantically similar items from a vector store.
